CODEBOOK FOR GRI STUDY DATA
================================

OVERVIEW:
This codebook documents variables from a field experiment studying discrimination against transgender individuals in India, conducted using a delivery service hiring task. The study implements a contact theory intervention: participants select delivery workers, receive an actual delivery from their chosen worker, and are surveyed again to measure attitude change. The experiment uses a group discussion design to study social influence on discriminatory behavior.

Data Files:
- GRI Main v2_WIDE.csv (Phase 1 Main Survey) - 1,181 variables
- GRI Phase 2_WIDE.csv (Phase 2 Main Survey) - 2,704 variables
- GRI Follow Up Delivery v3_WIDE.csv (Phase 1 Follow-up Survey) - 368 variables
- GRI Follow Up Delivery - Phase 2_WIDE.csv (Phase 2 Follow-up Survey) - 269 variables

STUDY DESIGN:
The main survey includes: (1) demographics and screening, (2) video intervention about rights (randomized to control/placebo/treatment), (3) hiring decisions between delivery workers varying in gender identity (cisgender vs transgender), education, reliability, and English proficiency, (4) group discussions (for group condition), (5) attitude and knowledge measures. Phase 2 introduces experimental enhancements including anonymous choice methods, announcement timing variation, and silent "choosing only" conditions.

The follow-up survey (administered after delivery) measures: (1) delivery experience, (2) post-intervention hiring preferences, (3) attitude and belief changes, (4) peer influence perception changes.

VARIABLE NAMING CONVENTIONS:
- Variables ending in "_label" contain labeled/text versions of coded responses
- Variables ending in "_other" contain open-ended text when "Other" is selected
- Variables with "_[1-n]" patterns represent indexed items (e.g., hiring pair 1, 2, 3...)
- Duration variables (duration_*) measure time in seconds spent on survey sections
- Indicator variables typically use 1=Yes, 0=No unless otherwise specified
- Missing/special codes: -99=Refuse to respond, -98=Don't know, -97=Other (specify), -96=None

================================================================================
GRI Main v2_WIDE.csv (Phase 1 Main Survey)
================================================================================

SURVEY METADATA AND TECHNICAL VARIABLES:
These variables track the survey administration process, device information, and data quality. They enable linking responses across survey waves and monitoring fieldwork quality.

SubmissionDate: Date and time when the survey form was submitted to the server

starttime: Date and time when the survey was started by the enumerator

endtime: Date and time when the survey was completed

deviceid: Unique identifier for the device (tablet) used to conduct the survey

subscriberid: Subscriber ID of the SIM card in the device

simid: SIM card identifier

devicephonenum: Phone number associated with the device

caseid: Unique case identifier assigned to each survey response

duration: Total duration of the survey in seconds

survey_date: Date when the survey was conducted

start_time_local: Local time when the survey started

supervisor: Supervisor ID code. Options: select_one from supervisor list

surveyor_id: Surveyor/enumerator ID code. Options: select_one from surveyor list

surveyor_id_other: Text entry for surveyor ID if "Other" is selected

surveyor_name: Full name of the surveyor/enumerator conducting the interview

team_id: Numeric team identifier assigned to the survey team

team_id_leading_zeros: Team ID formatted with leading zeros for display purposes

team_id_confirm: Confirmation of team ID. Options: 1=Yes, 0=No

district: District where the survey was conducted. Options: select_one from district list or -97=Other (specify)

district_other: Text entry for district name if "Other" is selected

district_lab: Processed label for district (alternative coding)

sector: Sector/area within the district where survey was conducted. Options: select_one from sector list or -97=Other (specify)

sector_other: Text entry for sector name if "Other" is selected

sector_lab: Processed label for sector (alternative coding)

location_other: Text entry for additional location details

location_lab: Processed label for location

group_id: Unique identifier for the group of respondents being interviewed together

id_exists: Indicator of whether the group ID exists in the database (1=exists, 0=does not exist)

id_check: Indicator of whether the group ID matches the team ID (1=match, 0=no match)

target_f: Numeric code for target gender (0=male, 1=female)

target_female: Binary indicator for whether target respondent should be female (0=male, 1=female)

group_size: Number of people in the group being interviewed

group_names_count: Count of group member names entered

group_person_i_[1-3]: Loop index for group member [1-3] used during data collection

group_name_[1-3]: Text entry for the name of group member [1-3]. Pattern repeats for up to 3 group members

group_person_[1-3]: Stored name of group person [1-3] for use in survey logic. Pattern repeats for up to 3 group members

group_check: Confirmation that group member names are in the same order across all tablets. Options: 1=Yes, 0=No

group_member_interviewed: Which group member will be interviewed using this tablet. Options: 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, -98=Don't know

duration_group_ids: Duration in seconds spent on the group IDs section

INDIVIDUAL AND TREATMENT ASSIGNMENT:
These variables identify individual respondents and their random assignment to experimental conditions (video type, group vs individual, delivery incentive structure).

ind_id: Individual respondent ID number

ind_id_error: Error indicator for individual ID validation

group: Indicator for whether respondent is part of a group interview (1=group, 0=individual)

video_type: Type of video shown to respondent. Options: control, placebo, or treatment

group_video: Video type assigned to the group

delivery_incentive: Type of delivery incentive offered. Options: single or multi

stratum_id: Stratification ID for sampling design

stratum_block: Block identifier within stratum for sampling

ind_name: Name of the individual respondent being interviewed

DEMOGRAPHICS AND SCREENING:
These variables collect respondent demographic information and implement screening criteria (age, literacy). They also serve as control variables in analysis.

b1: Age of respondent in years (integer)

b1_fail: Enumerator confirmation when age is outside required range (15-19 or >65). Options: 1=Yes (age correct), 0=No (go back and correct)

b8.1: Languages the respondent can speak. Options (select_multiple): 1=Tamil, 2=Kannada, 3=Telugu, 4=Malayalam, 5=Hindi, 6=English, -97=Other (specify), -96=No languages, -99=Refuse to respond

b8.1_other: Text entry for other languages spoken if -97 selected

b8.2: Languages the respondent can read a newspaper article in. Options (select_multiple): 1=Tamil, 2=Kannada, 3=Telugu, 4=Malayalam, 5=Hindi, 6=English, -97=Other (specify), -96=No languages, -99=Refuse to respond

b8.2_other: Text entry for other languages can read if -97 selected

b8.2_fail: Enumerator confirmation when respondent cannot read Tamil. Options: 1=Yes (correct, cannot read Tamil), 0=No (go back and correct)

b2: Gender of respondent (enumerator observe). Options: 1=Male, 2=Female, -97=Other (specify), -99=Refuse to respond

b2_other: Text entry for gender if -97 (Other) is selected

b2_fail: Enumerator confirmation when gender doesn't match target gender for this ID. Options: 1=Yes (correct, different from target), 0=No (go back and correct)

demographics_pass: Indicator of whether respondent passes demographic screening criteria (1=pass, 0=fail)

consent: Whether participant consented to participate in the study. Options: 1=Yes, 0=No

duration_consent: Duration in seconds spent on consent section

b3: Religion of respondent (enumerator observe if possible, otherwise ask). Options: 1=Hindu, 2=Christian, 3=Muslim, 4=Buddhist, 5=Jain, -97=Other (specify), -99=Refuse to respond

b3_other: Text entry for religion if -97 (Other) is selected

b4: Highest level of education completed. Options: 1=No education, 2=Primary school, 3=Middle school, 4=High School, 5=Higher Secondary school, 6=Bachelor's degree, 7=Master's degree, 8=ITI/Diploma, -97=Other (specify), -99=Refuse to respond

b4_other: Text entry for education level if -97 (Other) is selected

b5: Current marital status. Options: 1=Married, 2=Unmarried, 3=Divorced, 4=Separated, 5=Widower/Widow, -97=Other (specify), -99=Refuse to respond

b5_other: Text entry for marital status if -97 (Other) is selected

b6: Current employment status. Options: 1=Not employed, 2=Part-time employed, 3=Full-time employed, 4=Seasonally employed, 5=Housewife/homemaker, -97=Other (specify), -99=Refuse to respond

b6_other: Text entry for employment status if -97 (Other) is selected

b6.1: Whether respondent owns any housing that they rent out. Options: 1=Yes, 0=No, -99=Refuse to respond

b7: Number of children aged 18 or below living in respondent's household (integer)

b8: Whether respondent has hired someone to work for them in the last 2 years (as part of occupation or at home). Options: 1=Yes, 0=No, -99=Refuse to respond

b11: Whether respondent has ever taken part in a market research survey by a company. Options: 1=Yes, 0=No, -99=Refuse to respond

b12: Whether respondent has ever received a free item as part of a promotion. Options: 1=Yes, 0=No, -99=Refuse to respond

b13: Number of people in respondent's household (integer)

b14: Approximate monthly household spending on food in rupees (integer)

duration_demographics: Duration in seconds spent on demographics section

TECHNOLOGY USE AND DELIVERY FAMILIARITY:
These questions assess household familiarity with app-based delivery services, establishing baseline exposure to the delivery economy context.

i1: Whether anyone in household has ever used an app/website to order a taxi/motorbike/auto. Options: 1=Yes, 0=No, -98=Don't know, -99=Refuse to respond

i2: Whether anyone in household has ever used an app/website to order food, drink, or groceries. Options: 1=Yes, 0=No, -98=Don't know, -99=Refuse to respond

i3: Whether anyone in household has ever used an app/website to order other goods (such as clothing, items for the house). Options: 1=Yes, 0=No, -98=Don't know, -99=Refuse to respond

duration_delivery_questions: Duration in seconds spent on delivery questions section

SOCIAL DESIRABILITY BIAS SCALE:
This validated scale measures respondents' tendency to provide socially desirable responses. Used to control for social desirability bias in attitudes toward transgender individuals.

sdb_2: Social desirability statement: "I sometimes feel annoyed at people when I don't get what I want". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

sdb_5: Social desirability statement: "No matter who I'm talking to, I'm always a good listener". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

sdb_8: Social desirability statement: "I sometimes try to take revenge instead of forgiving and forgetting". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

sdb_9: Social desirability statement: "I am always polite, even to people who are not nice". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

sdb_11: Social desirability statement: "There have been times when I was jealous of other people's luck". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

sdb_12: Social desirability statement: "I am sometimes annoyed when people ask me for favours". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

sdb_13: Social desirability statement: "I have deliberately said something that hurt someone's feelings". Options: 1=Agree, 2=Disagree, -99=Refuse to answer

duration_sdb: Duration in seconds spent on social desirability bias section

ITEM PREFERENCE AND DELIVERY SERVICE QUESTIONS:
These questions establish preferences for household goods and willingness to pay for delivery, making the hiring task more concrete and realistic.

dq1: Item preference: Which item do you prefer? Options: 1=Sunflower oil (1l), 2=Groundnut oil (1l), -99=Refuse to answer

dq3: Item preference: Which item do you prefer? Options: 1=Horlicks (500g), 2=Boost (500g), -99=Refuse to answer

duration_delivery_qs_b1: Duration in seconds spent on delivery questions (b1) section

dq10: Logo preference for delivery service. Options: 1=Snappy, 2=Sprint, -99=Refuse to answer

dq12: Additional amount willing to pay for delivery to home in rupees (integer)

duration_delivery_qs_b2: Duration in seconds spent on delivery questions (b2) section

PRACTICE TASK:
This section familiarizes respondents with the pairwise choice task and scratchcard incentive mechanism before the actual hiring decisions. It ensures comprehension and comfort with the task structure.

n_practice_items: Number of practice item pairs to be shown (4)

n_practice_items_less1: Number of practice items minus 1 (3)

n_practice_items_double: Number of practice items times 2 (8)

d1: Practice item selection: Which item would you prefer? Options: 1=Clinic-Plus Shampoo, 2=Head and shoulders Shampoo, -99=Refuse to answer

practice_check_2: Comprehension check: How many items will you have selected by the end of the game? (integer, correct answer=4)

practice_check_3: Comprehension check: How many items will you actually receive? (integer, correct answer=1)

practice_check_7: Comprehension check: How will it be decided which one item you actually receive? Options: 1=Using the scratchcards, 0=Other incorrect answer, -98=Don't know

practice_check_4: Comprehension check: Did you understand the process now? Options: 1=Yes, 0=No, -98=Don't know

d2: Practice item selection: Which item would you prefer? Options: 1=Milk Bikis, 2=Sunrise Coffee, -99=Refuse to answer

d3: Practice item selection: Which item would you prefer? Options: 1=Parle-G, 2=Sunrise Coffee, -99=Refuse to answer

d4: Practice item selection: Which item would you prefer? Options: 1=Ariel, 2=Milk Bikis, -99=Refuse to answer

d4_diversity_check_after: Why did you choose this item? (shown if d2=1 and d4=1 to check if respondent is treating each pair independently). Options: 1=Because I already chose the other item before, -97=Other reason

practice_sc_number: Which number was scratched on the practice scratchcard? Options: 1=1.0, 2=2.0, 3=3.0, 4=4.0

practice_sc_refused: Indicator if respondent refused to select from a pair in practice (1=refused, 0=selected)

practice_refusal_photo_1: Photo/item identifier for first option in refused practice pair

practice_refusal_photo_2: Photo/item identifier for second option in refused practice pair

practice_refusal_choice_show: Display of the two items in the refused pair. Options: 1=Item 1, 2=Item 2

practice_refusal_choice_show_yn: Does respondent want to choose now? Options: 1=Yes, 0=No

practice_refusal_choice: Which item does respondent choose (if they agree to choose)? Options: 1=Item 1, 2=Item 2, -99=Refuse to answer

practice_refusal_coin_yn: Does respondent want to flip a coin to decide (if they don't want to choose)? Options: 1=Yes, 0=No

duration_practice: Duration in seconds spent on practice section

HIRING TASK INSTRUCTIONS AND COMPREHENSION CHECKS:
These variables implement comprehension checks to ensure respondents understand the hiring task: they will evaluate multiple delivery worker pairs, considering worker characteristics and items to be delivered, and one choice will be randomly selected for implementation.

n_hiring_pairs: Total number of hiring decision pairs (10)

n_hiring_pairs_r1: Number of hiring pairs in round 1 (4)

n_hiring_pairs_r1_lessone: Number of hiring pairs in round 1 minus 1 (3)

n_hiring_pairs_r1_plusone: Number of hiring pairs in round 1 plus 1 (5)

n_hiring_pairs_r1_plustwo: Number of hiring pairs in round 1 plus 2 (6)

n_hiring_pairs_r2: Number of hiring pairs in round 2 (6)

n_deliveries: Number of deliveries (1 for single incentive, 3 for multi incentive)

n_deliveries_lessone: Number of deliveries minus 1

hiring_check_1: Comprehension check: Which three things do you need to consider? Options (select_multiple): 1=Person and their details, 2=Items, 3=Conversation for 15 minutes, 0=None, -98=Don't know

hiring_check_1_pass_cond: Indicator of whether respondent passed hiring check 1 (1=pass, 0=fail)

hiring_check_5: Comprehension check: In total, how many delivery options will you have to choose? (integer, correct answer=10)

hiring_check_3: Comprehension check: How many deliveries will you receive? (integer, correct answer depends on delivery_incentive)

hiring_check_5.1: Comprehension check: How will we select which option you actually receive? Options: 1=Using the scratchcards, 0=Other incorrect answer, -98=Don't know

hiring_check_11: Comprehension check: Do you understand the process? Options: 1=I understand the instructions, 2=I would like you to explain again

duration_delivery_choice_instructions: Duration in seconds spent on delivery choice instructions section

VIDEO INTERVENTION:
Respondents watch one of three videos (randomized): (1) Control: consumer rights and voting rights, (2) Placebo: consumer rights, voting rights, and transgender rights framed as aspirational ("should have"), (3) Treatment: consumer rights, voting rights, and transgender rights framed as legal fact granted by Supreme Court ("do have"). Video recall questions measure attention and comprehension, enabling identification of the intervention's mechanism.

duration_before_video: Duration in seconds before video was shown

duration_video: Duration of video playback in seconds

c1: Video recall: What do you remember from the video? Options (select_multiple): 1=Rights, 2=Rights are rules about what people are entitled to, 3=Consumer rights, 4=When you make a purchase you have a right to complain if unfair, 21=Voting rights, 22=People have the right to vote, 23=In India anyone aged above 18 is entitled to vote and elect representatives, 31=Thirunangai, 32=Thirunangai rights (general), 33=Thirunangai housing rights, 34=Thirunangai employment rights, 35=Thirunangai education rights, 36=Thirunangai fundamental rights, 41=Supreme Court, 42=Supreme Court gave thirunangai rights, 43=Law gave thirunangai rights, 45=Supreme Court is most powerful legal institution, 51=Worker rights, 52=Delivery workers, 53=ESI/PF/Insurance, 54=Delivery workers have worker rights when you make a purchase, 55=Delivery workers think they should have right to employment benefits, 56=Delivery workers do have the right to employment benefits, -97=Other (specify), 0=Don't remember anything more, -99=Refuse to answer

c1_other: Text entry for other things remembered from video

c2: Video recall follow-up: What specifically did the video say about rights? (shown if selected c1=1 but not specific details). Options: same as c1

c2_other: Text entry for other details about rights from video

c7: Video recall follow-up: What specifically did the video say about worker rights? (shown if selected c1=22 but not other details). Options: same as c1

c7_other: Text entry for other details about worker rights from video

c4: Video recall follow-up: What specifically did the video say about voting rights? (shown if selected c1=21 but not other details). Options: same as c1

c4_other: Text entry for other details about voting rights from video

c5: Video recall follow-up: What specifically did the video say about thirunangai rights? (shown if selected c1=31 or 32 but not other details). Options: same as c1

c5_other: Text entry for other details about thirunangai rights from video

c6: Video recall follow-up: What specifically did the video say about the Supreme Court? (shown if selected c1=41 but not other details). Options: same as c1

c6_other: Text entry for other details about Supreme Court from video

c_joined: Combined/processed variable joining all video recall responses

c_mentioned_thirunangai: Indicator of whether respondent mentioned thirunangai in video recall (1=mentioned, 0=not mentioned)

vid_should_do: Enumerator observation: Did respondent say thirunangai should have rights or they do have rights? Options (select_multiple): 1=They should have rights, 2=They do have rights, -99=Refuse to respond

vid_law_distinction: Enumerator observation: Did respondent mention the law or the Supreme Court? Options: 1=Yes, 0=No, -99=Refuse to respond

c17: Which type of workers did the video say think they should have a right to pension and insurance benefits? Options: 1=Delivery workers, 2=Farmers, 3=Construction workers, 4=Government workers, -98=Don't know, -99=Refuse to respond

c14: What did the video say about the rights of thirunangai? (shown for placebo video if thirunangai not mentioned). Options (select_multiple): 1=Rights, 2=Fundamental rights, 3=Equal rights/rights to equality, 5=Rights granted under Indian constitution, 6=Education, 7=Employment, 8=Housing, 9=Freedom from discrimination, 13=Right to vote, 10=They are also human, 11=They are equal/like us, 12=We should respect them, -97=Other (specify), -98=Don't know, -99=Refuse to respond

c14_other: Text entry for other details about thirunangai rights (placebo)

c15: What is the main role of the Supreme Court? (shown for treatment video). Options: 1=Maintain and protect the law, 2=Conduct the elections, 3=Construct roads, 4=Print money, -98=Don't know, -99=Refuse to respond

c16: What rights did the Supreme Court give to thirunangai people? (shown for treatment video). Options (select_multiple): same as c14

c16_other: Text entry for other rights given by Supreme Court

c16_mentioned_rights: Indicator of whether respondent mentioned rights in c16 (1=mentioned, 0=not mentioned)

c16_should_do: Enumerator observation: Did respondent say should have or do have rights in c16? Options (select_multiple): 1=They should have rights, 2=They do have rights, -99=Refuse to respond

video_pamphlet: Type of pamphlet prepared matching video type (control=Blue, treatment=Yellow, placebo=Pink)

duration_video_text_prep: Duration in seconds spent preparing video pamphlet

video_text_colour: Which colour pamphlet was prepared? Options: control=Blue, treatment=Yellow, placebo=Pink

duration_video_text_colour: Duration in seconds spent on pamphlet colour check

duration_video_text_colour_fail: Duration in seconds for failed colour check (wrong colour selected)

duration_video_text: Duration in seconds for giving pamphlet with video information

duration_video_read: Duration in seconds allowing respondent to read pamphlet (2 minutes)

duration_video_pamphlet_take_back: Duration in seconds for taking back pamphlet

duration_video_questions: Duration in seconds spent on video questions section

o1: Working conditions for Swiggy & Zomato workers - are they good or bad? Options: 1=Generally good, 2=Generally bad, 3=Mixed, -98=Don't know, -99=Refuse to respond

o2: How should companies improve safety for delivery workers? Options (select_multiple): 1=Protective equipment, 2=Training, 3=Insurance, 4=Limit working hours, 5=Increase salary, -97=Other (specify), -98=Don't know, -99=Refuse to respond

o2_other: Text entry for other safety improvement suggestions

duration_video_open_questions: Duration in seconds spent on open questions about video

GROUP DISCUSSION SECTION:
For respondents assigned to the group condition, this section implements group discussion about hiring choices. One group member acts as discussion leader, another records observations. Audio is recorded (with consent) for qualitative analysis. The group discussion design allows measurement of peer influence on discriminatory behavior.

include_audio: Indicator for whether audio recording is included (1=yes, 0=no)

duration_group_move: Duration in seconds for moving to group discussion area

group_role: Which role will this tablet be used for in group discussion? Options: 1=Discussion leader, 2=Observation notes, 5=Assistant

audio_consent_lead: Does the group consent to be recorded? (group leader tablet). Options: 1=Yes, 0=No

discuss_main_audio: Audio recording file name for main discussion

audio_consent_observer: Does the group consent to be recorded? (observer tablet). Options: 1=Yes, 0=No, -98=Don't know

o1_obs: Observer notes: Do they think working conditions for Swiggy & Zomato workers are good or bad? Options: 1=Generally good, 2=Generally bad, 3=Mixed, -98=Don't know, -99=Refuse to respond

o2_obs: Observer notes: How do they think companies should improve safety for delivery workers? Options (select_multiple): 1=Protective equipment, 2=Training, 3=Insurance, 4=Limit working hours, 5=Increase salary, -97=Other (specify), -98=Don't know, -99=Refuse to respond

o2_obs_other: Text entry for other safety improvement suggestions (observed)

gobs_video_general1: In general, how much discussion was there about delivery workers? Options: 0=No discussion (only choice made), 1=Very little discussion, 2=Some discussion, 3=Lots of discussion

discuss_audio_audit: Audio recording file name for hiring discussion

group_repeat_pair_count: Number of hiring pairs observed in group discussion (up to 4)

group_pair_i_[1-4]: Loop index for hiring pair [1-4] being observed. Pattern repeats for up to 4 pairs

group_obs_index_1_[1-4]: Index of first delivery worker in observed hiring pair [1-4]. Pattern repeats for up to 4 pairs

group_obs_index_2_[1-4]: Index of second delivery worker in observed hiring pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_obs_trans_1_[1-4]: Indicator if first worker in observed pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

r1_discuss_obs_trans_2_[1-4]: Indicator if second worker in observed pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

group_obs_trans_included_[1-4]: Indicator if a transgender worker is included in observed pair [1-4] (1=included, 0=not). Pattern repeats for up to 4 pairs

gobs1_[1-4]: Who spoke first in discussion of pair [1-4]? Options: 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, -98=Don't know. Pattern repeats for up to 4 pairs

gobs2_[1-4]: Who is leading the conversation for pair [1-4]? Options (select_multiple): 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, -98=Don't know. Pattern repeats for up to 4 pairs

gobs3_[1-4]: What is the main reason(s) for the choice in pair [1-4]? Options (select_multiple): 1=It's a lady, 2=It's a transgender, 3=It's a man, 4=I wanted this item/these items or items are better, 5=They are also human, 6=I want to give opportunity or help them, 7=Person seems poor, 8=Easy to talk with this person, 9=I can relate easily, 10=Person seemed reliable/trustworthy, 11=Person seemed friendly, 12=Person was good looking, 13=Person is older, 14=Person is younger, 15=Person is more educated, 16=Person is less educated, 17=Higher reliability score/more experienced, 18=Lower reliability score/less experienced, 19=They speak English, -97=Other (specify), -98=Don't know, -99=Refuse to respond. Pattern repeats for up to 4 pairs

gobs6_[1-4]: How much discussion occurred for pair [1-4]? Options: 0=No discussion (only choice made), 1=Very little discussion, 2=Some discussion, 3=Lots of discussion. Pattern repeats for up to 4 pairs

gobs5_[1-4]: Did someone say something positive about transgenders in pair [1-4]? (shown if transgender included). Options: 1=Yes, 0=No, -98=Don't know. Pattern repeats for up to 4 pairs

gobs5.2_[1-4]: Did someone say something negative about transgenders in pair [1-4]? (shown if transgender included). Options: 1=Yes, 0=No, -98=Don't know. Pattern repeats for up to 4 pairs

gobs5.1_[1-4]: Who spoke in favour of transgenders in pair [1-4]? (shown if transgender included). Options (select_multiple): 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, 0=No-one, -98=Don't know. Pattern repeats for up to 4 pairs

gobs3_other_[1-4]: Text entry for other reasons given in pair [1-4]. Pattern repeats for up to 4 pairs

NOTE: Variables 301-347 continue the pattern from variables 285-300, covering observation data for hiring pairs 2, 3, and 4 in group discussions. See descriptions for variables with [1-4] pattern above for details on each variable type.

gobs7: In general, who was the most dominant in the conversation and group decision? Options: 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, -98=Don't know

gobs8: In general, how much discussion occurred across all hiring pairs? Options: 0=No discussion (only choice made), 1=Very little discussion, 2=Some discussion, 3=Lots of discussion

group_person_att_count: Number of group persons whose attitude towards transgenders was observed (3)

group_person_att_i_[1-3]: Loop index for group person [1-3] attitude observation. Pattern repeats for 3 group members

gobs9_[1-3]: In general, what was ${group_person_att_i}'s attitude towards transgenders? Options: 1=Strongly against choosing transgenders, 2=Against choosing transgenders, 3=Pro choosing transgenders, 4=Strongly pro choosing transgenders, 0=Did not discuss transgenders, -98=Don't know. Pattern repeats for 3 group members

other_obs: Text entry for any other observations about the group discussion

r1_check_photo_1: Photo filename for first worker shown in verification check

r1_check_educ_1: Education level of first worker shown (for verification)

r1_check_reliability_1: Reliability score of first worker shown (for verification)

r1_check_english_1: English proficiency of first worker shown (for verification)

r1_check_items_1: Items to be delivered by first worker shown (for verification)

r1_check_trans_1: Indicator if first worker shown is transgender (1=transgender, 0=not)

r1_check_photo_2: Photo filename for second worker shown in verification check

r1_check_educ_2: Education level of second worker shown (for verification)

r1_check_reliability_2: Reliability score of second worker shown (for verification)

r1_check_english_2: English proficiency of second worker shown (for verification)

r1_check_items_2: Items to be delivered by second worker shown (for verification)

r1_check_trans_2: Indicator if second worker shown is transgender (1=transgender, 0=not)

group_choice_check_show: Display of first choice on sheets for verification. Options: 1=Person A, 2=Person B

group_choice_check_yn: Do the choice sheets match the options shown? Options: 1=Yes, 0=No

group_repeat_count: Number of group hiring decision pairs (up to 4)

group_i_[1-4]: Loop index for group hiring decision pair [1-4]. Pattern repeats for up to 4 pairs

group_index_1_[1-4]: Index of first delivery worker in group pair [1-4]. Pattern repeats for up to 4 pairs

group_index_2_[1-4]: Index of second delivery worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_photo_1_[1-4]: Photo filename for first worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_educ_1_[1-4]: Education level of first worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_reliability_1_[1-4]: Reliability score of first worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_english_1_[1-4]: English proficiency of first worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_items_1_[1-4]: Items to be delivered by first worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_trans_1_[1-4]: Indicator if first worker in group pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

r1_discuss_photo_2_[1-4]: Photo filename for second worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_educ_2_[1-4]: Education level of second worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_reliability_2_[1-4]: Reliability score of second worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_english_2_[1-4]: English proficiency of second worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_items_2_[1-4]: Items to be delivered by second worker in group pair [1-4]. Pattern repeats for up to 4 pairs

r1_discuss_trans_2_[1-4]: Indicator if second worker in group pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

group_choice_[1-4]: Which delivery option did the group choose for pair [1-4]? Options: 1=Person A, 2=Person B, -99=Refuse to answer. Pattern repeats for up to 4 pairs

group_choice_backup_[1-4]: Backup entry of which option the group selected in pair [1-4] (from non-assistant tablet). Options: 1=Person A, 2=Person B, -99=Refuse to answer. Pattern repeats for up to 4 pairs

NOTE: Variables 401-450 consist primarily of the specific instances for hiring pairs 2, 3, and 4 (indices _2, _3, _4) of the patterns described above, following the same structure as pair 1

group_[1-4]: Final group choice for hiring pair [1-4] (1 or 2). Pattern repeats for up to 4 pairs

hiring_photo_r1_group_final_[1-8]: Photo filename for the [1-8] workers selected by groups in round 1. Pattern repeats for up to 8 workers

duration_group_section: Duration in seconds spent on group hiring decision section

ind_repeat_count: Number of individual hiring decision pairs (varies by condition)

r1_no_discuss_i_[1-4]: Loop index for individual (non-group) hiring decision pair [1-4] in round 1. Pattern repeats for up to 4 pairs

r1_no_discuss_index_1_[1-4]: Index of first delivery worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_index_2_[1-4]: Index of second delivery worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_photo_1_[1-4]: Photo filename for first worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_age_1_[1-4]: Age of first worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_educ_1_[1-4]: Education level of first worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_reliability_1_[1-4]: Reliability score of first worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_english_1_[1-4]: English proficiency of first worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_items_1_[1-4]: Items to be delivered by first worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_trans_1_[1-4]: Indicator if first worker in individual pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

r1_no_discuss_photo_2_[1-4]: Photo filename for second worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_age_2_[1-4]: Age of second worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_educ_2_[1-4]: Education level of second worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_reliability_2_[1-4]: Reliability score of second worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_english_2_[1-4]: English proficiency of second worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_items_2_[1-4]: Items to be delivered by second worker in individual pair [1-4]. Pattern repeats for up to 4 pairs

r1_no_discuss_trans_2_[1-4]: Indicator if second worker in individual pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

ind_choice_[1-4]: Which delivery option did the individual choose for pair [1-4]? Options: 1=Person A, 2=Person B, -99=Refuse to answer. Pattern repeats for up to 4 pairs

ind_choice_refusal_reason_[1-4]: Reason for refusing to choose in individual pair [1-4] (if refused). Pattern repeats for up to 4 pairs

NOTE: Variables 451-500 continue the pattern for individual hiring pairs 2, 3, and 4, following the same structure as pair 1

NOTE: Variables 501-550 consist of the specific instances for individual hiring pairs 2, 3, and 4 (indices _2, _3, _4), continuing the same structure

duration_hiring_round_1: Duration in seconds spent on round 1 hiring decisions

sc_colour: Color of the scratchcard used. Options: 1=Red, 2=Blue, 3=Green, 4=Yellow, 5=Pink

sc_colour_check: Enumerator check that correct scratchcard color was selected. Options: 1=Correct, 0=Incorrect

duration_scratchcard_choice: Duration in seconds for scratchcard selection process

check_private: Confirmation that respondent is in a private location for sensitive questions (group interviews). Options: 1=Yes, 0=No

check_private_ind: Confirmation that respondent is in a private location for sensitive questions (individual interviews). Options: 1=Yes, 0=No

hiring_repeat_count: Number of hiring decision pairs in round 2 (6 pairs)

r2_i_[1-6]: Loop index for round 2 hiring decision pair [1-6]. Pattern repeats for up to 6 pairs

r2_index_1_[1-6]: Index of first delivery worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_index_2_[1-6]: Index of second delivery worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

hiring_i_for_r2_[1-6]: Overall hiring index for round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_photo_1_[1-6]: Photo filename for first worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_educ_1_[1-6]: Education level of first worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_reliability_1_[1-6]: Reliability score of first worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_english_1_[1-6]: English proficiency of first worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_items_1_[1-6]: Items to be delivered by first worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_trans_1_[1-6]: Indicator if first worker in round 2 pair [1-6] is transgender (1=transgender, 0=not). Pattern repeats for up to 6 pairs

r2_photo_2_[1-6]: Photo filename for second worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_educ_2_[1-6]: Education level of second worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_reliability_2_[1-6]: Reliability score of second worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_english_2_[1-6]: English proficiency of second worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_items_2_[1-6]: Items to be delivered by second worker in round 2 pair [1-6]. Pattern repeats for up to 6 pairs

r2_trans_2_[1-6]: Indicator if second worker in round 2 pair [1-6] is transgender (1=transgender, 0=not). Pattern repeats for up to 6 pairs

hiring_choice_[1-6]: Which delivery option did respondent choose in round 2 pair [1-6]? Options: 1=Person A, 2=Person B, -99=Refuse to answer. Pattern repeats for up to 6 pairs

hiring_choice_refusal_reason_[1-6]: Reason for refusing to choose in round 2 pair [1-6] (if refused). Pattern repeats for up to 6 pairs

NOTE: Variables 551-600 continue the pattern for round 2 hiring pairs 2 and beyond, following the same structure as pair 1

NOTE: Variables 601-650 consist of the specific instances for round 2 hiring pairs 2, 3, 4, and 5 (indices _2, _3, _4, _5), continuing the same structure

hired_[1-6]: Final hiring decision for pair [1-6] (1 or 2). Pattern repeats for up to 6 pairs

hired_refused_[1-6]: Indicator of whether respondent refused to hire for pair [1-6] (1=refused, 0=made choice). Pattern repeats for up to 6 pairs

duration_hiring: Total duration in seconds spent on all hiring decisions

f1: Reason for hiring choices. Options: (select_one or select_multiple - details from questionnaire)

f1_other: Text entry for other reasons if -97 (Other) is selected in f1

worker_rating_1_photo: Photo filename for first worker being rated

worker_rating_2_photo: Photo filename for second worker being rated

worker_rating_switch: Indicator of whether worker order was switched for rating (to avoid order effects)

worker_rating_1_photo_actual: Actual photo filename for first worker after accounting for switch

worker_rating_2_photo_actual: Actual photo filename for second worker after accounting for switch

f7.1: Rating or comparison question for worker 1

f7.2: Rating or comparison question for worker 2

f3: Follow-up question about hiring decisions

f4: Follow-up question about hiring decisions

f4_other: Text entry for other responses if -97 (Other) is selected in f4

f5.1: Sub-question related to hiring decisions

f5.2: Sub-question related to hiring decisions

f5.3: Sub-question related to hiring decisions

f6: Follow-up question about hiring decisions

f2.0: Question about hiring decisions or reasoning

f2.1: Question about hiring decisions or reasoning

f2.1_single: Single delivery version of f2.1 question

f2.4: Question about hiring decisions or reasoning

duration_reasons_choices: Duration in seconds spent on reasons for hiring choices section

ind_[1-4]: Final individual choice for hiring pair [1-4] (1 or 2). Pattern repeats for up to 4 pairs

hiring_photo_r1_ind_final_[1-8]: Photo filename for the [1-8] workers selected by individuals in round 1. Pattern repeats for up to 8 workers

hiring_items_r1_ind_final_[1-8]: Items to be delivered by the [1-8] workers selected by individuals in round 1. Pattern repeats for up to 8 workers

hiring_items_r1_group_final_[1-8]: Items to be delivered by the [1-8] workers selected by groups in round 1. Pattern repeats for up to 8 workers

hiring_items_text_r1_ind_final_[1-8]: Text description of items for the [1-8] workers selected by individuals in round 1. Pattern repeats for up to 8 workers

hiring_items_text_r1_group_final_[1-8]: Text description of items for the [1-8] workers selected by groups in round 1. Pattern repeats for up to 8 workers

hiring_photo_final_[1-14]: Photo filename for the [1-14] workers selected across all rounds (both round 1 and round 2). Pattern repeats for up to 14 workers

hiring_items_final_[1-14]: Items to be delivered by the [1-14] workers selected across all rounds. Pattern repeats for up to 14 workers

hiring_items_text_final_[1-14]: Text description of items for the [1-14] workers selected across all rounds. Pattern repeats for up to 14 workers

sc_number: Number revealed on the scratchcard that determines which hiring choice is implemented (integer 1-10)

real_trans_delivery: Indicator of whether the selected delivery worker for implementation is transgender (1=transgender, 0=not transgender)

hiring_selected_round: Which round was selected via scratchcard for implementation (1=round 1, 2=round 2)

hiring_selected_choice: Which hiring pair choice was selected via scratchcard for implementation (integer)

hiring_selected_photo_1: Photo filename for first worker in the selected delivery pair

hiring_selected_educ_1: Education level of first worker in the selected delivery pair

hiring_selected_reliability_1: Reliability score of first worker in the selected delivery pair

hiring_selected_english_1: English proficiency of first worker in the selected delivery pair

hiring_selected_items_1: Items to be delivered by first worker in the selected delivery pair

hiring_selected_photo_2: Photo filename for second worker in the selected delivery pair

hiring_selected_educ_2: Education level of second worker in the selected delivery pair

hiring_selected_reliability_2: Reliability score of second worker in the selected delivery pair

hiring_selected_english_2: English proficiency of second worker in the selected delivery pair

hiring_selected_items_2: Items to be delivered by second worker in the selected delivery pair

deliv_refusal_choice_show: Display of the two workers in the selected pair if respondent initially refused

deliv_refusal_choice_show_yn: Does respondent want to choose now between the two workers? Options: 1=Yes, 0=No

deliv_refusal_choice: Which worker does respondent choose (if they agree to choose)? Options: 1=Worker 1, 2=Worker 2, -99=Refuse to answer

deliv_refusal_coin_yn: Does respondent want to flip a coin to decide (if they don't want to choose)? Options: 1=Yes, 0=No

deliv_refusal_coin_flip: Result of coin flip if respondent agreed to coin flip. Options: 1=Worker 1, 2=Worker 2

hiring_selected_choice_extra: Additional processing variable for selected choice

hiring_selected_photo: Photo filename of the worker ultimately selected for delivery

hiring_selected_items: Items to be delivered by the worker ultimately selected for delivery

team_id_for_delivery: Team ID to be used for coordinating the delivery

enum_id_for_delivery: Enumerator ID to be used for coordinating the delivery

enum_name_for_delivery: Enumerator name to be used for coordinating the delivery

receives_delivery: Indicator of whether respondent will receive delivery (1=yes, 0=no)

deliv_option_[1-5]: Delivery option [1-5] details (for multi-delivery condition). Pattern repeats for up to 5 deliveries

dq9: What motivated your hiring choices? Options: (select_one or select_multiple from questionnaire)

dq9_other: Text entry for other motivations if -97 (Other) is selected in dq9

delivery_days: Which day(s) would you like to receive delivery? Options: (select_one or select_multiple from questionnaire)

delivery_times: What time would you like to receive delivery? Options: (select_one or select_multiple from questionnaire)

delivery_times_other: Text entry for other time preferences if -97 (Other) is selected

delivery_times_lab_[1-4]: Processed labels for delivery time preferences [1-4]. Pattern repeats for up to 4 time slots

delivery_times_lab: Combined processed label for delivery times

address: Delivery address provided by respondent (text)

phone: Phone number provided by respondent for delivery coordination (text)

duration_finalise_delivery: Duration in seconds spent finalizing delivery details

purpose_0: What do you think was the main purpose of this survey? Options: (select_one or select_multiple from questionnaire)

purpose_0_other: Text entry for other perceived purposes if -97 (Other) is selected

duration_perceived_purpose: Duration in seconds spent on perceived purpose question

dd2: Discussion/debrief question about delivery workers

dd3: Discussion/debrief question about delivery workers

dd4.1: Discussion/debrief question about delivery workers

dd5: Discussion/debrief question about delivery workers

dd7: Discussion/debrief question about delivery workers

dd8: Discussion/debrief question about delivery workers

dd9: Discussion/debrief question about delivery workers

duration_discussion_debrief: Duration in seconds spent on discussion/debrief section

k1: Knowledge or awareness question. Options: (from questionnaire)

k1_other: Text entry for other responses if -97 (Other) is selected in k1

k2: Knowledge or awareness question. Options: (from questionnaire)

k2.2: Follow-up knowledge question

k2.2_other: Text entry for other responses if -97 (Other) is selected in k2.2

k2.2_should_do: Enumerator observation: Did respondent say should have or do have rights in k2.2? Options (select_multiple): 1=They should have rights, 2=They do have rights, -99=Refuse to respond

k3: Knowledge or awareness question. Options: (from questionnaire)

k4: Knowledge or awareness question. Options: (from questionnaire)

k5: Knowledge or awareness question. Options: (from questionnaire)

k6: Knowledge or awareness question. Options: (from questionnaire)

k6_other: Text entry for other responses if -97 (Other) is selected in k6

k6.1: Follow-up knowledge question

k6.2: Follow-up knowledge question

k6.3: Follow-up knowledge question

k7: Knowledge or awareness question. Options: (from questionnaire)

k8: Knowledge or awareness question. Options: (from questionnaire)

k8.2: Follow-up knowledge question

k9: Knowledge or awareness question. Options: (from questionnaire)

k10: Knowledge or awareness question. Options: (from questionnaire)

k11: Knowledge or awareness question. Options: (from questionnaire)

duration_beliefs_law: Duration in seconds spent on beliefs/law questions section

norms_[1-6]_photo: Photo filename for worker [1-6] shown in social norms questions. Pattern repeats for up to 6 workers

norms_[1-6]_items: Items to be delivered by worker [1-6] shown in social norms questions. Pattern repeats for up to 6 workers

norms_right: Indicator or value for the "right" side in norms questions

n1_show_right: Indicator of whether option shown on right side for norms question 1. Options: 1=shown on right, 0=shown on left

n1_right: Response to norms question 1 about the option shown on the right

n2_show_right: Indicator of whether option shown on right side for norms question 2. Options: 1=shown on right, 0=shown on left

n2_right: Response to norms question 2 about the option shown on the right

n3_show_right: Indicator of whether option shown on right side for norms question 3. Options: 1=shown on right, 0=shown on left

n3_right: Response to norms question 3 about the option shown on the right

n1_show_left: Indicator of whether option shown on left side for norms question 1. Options: 1=shown on left, 0=shown on right

n1_left: Response to norms question 1 about the option shown on the left

n2_show_left: Indicator of whether option shown on left side for norms question 2. Options: 1=shown on left, 0=shown on right

n2_left: Response to norms question 2 about the option shown on the left

n3_show_left: Indicator of whether option shown on left side for norms question 3. Options: 1=shown on left, 0=shown on right

n3_left: Response to norms question 3 about the option shown on the left

duration_social_norms: Duration in seconds spent on social norms section

group_predic_member_a: First group member selected for prediction task

group_predic_member_b: Second group member selected for prediction task

group_predic_name_a: Name of first group member for prediction task

group_predic_name_b: Name of second group member for prediction task

group_predic_random_draw: Random assignment for prediction task order

group_predic_which_first: Which group member prediction comes first (a or b)

group_predic_repeat_count: Number of prediction pairs to complete

group_predic_i_[1-4]: Loop index for group prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_a_or_b_[1-4]: Whether predicting for member a or b in pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_index_1_[1-4]: Index of first delivery worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_index_2_[1-4]: Index of second delivery worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_photo_1_[1-4]: Photo filename for first worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_age_1_[1-4]: Age of first worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_educ_1_[1-4]: Education level of first worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_reliability_1_[1-4]: Reliability score of first worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_english_1_[1-4]: English proficiency of first worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_items_1_[1-4]: Items to be delivered by first worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_trans_1_[1-4]: Indicator if first worker in prediction pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

group_predic_photo_2_[1-4]: Photo filename for second worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_age_2_[1-4]: Age of second worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_educ_2_[1-4]: Education level of second worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_reliability_2_[1-4]: Reliability score of second worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_english_2_[1-4]: English proficiency of second worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_items_2_[1-4]: Items to be delivered by second worker in prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_trans_2_[1-4]: Indicator if second worker in prediction pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

group_predic_name_[1-4]: Name of group member being predicted for in pair [1-4]. Pattern repeats for up to 4 pairs

group_member_relation_[1-4]: Relationship to group member being predicted for in pair [1-4]. Pattern repeats for up to 4 pairs

group_member_relation_other_[1-4]: Text entry for other relationship if -97 (Other) selected. Pattern repeats for up to 4 pairs

group_member_know_[1-4]: How well respondent knows group member in pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_choice_[1-4]: Predicted choice of group member in pair [1-4]. Options: 1=Person A, 2=Person B. Pattern repeats for up to 4 pairs

group_predic_refusal_reason_[1-4]: Reason for refusal if respondent refused to predict in pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_refusal_reason_other_[1-4]: Text entry for other refusal reason. Pattern repeats for up to 4 pairs

NOTE: Variables 951-1000 continue the pattern for group prediction pairs 2, 3, and 4, following the same structure as pair 1

NOTE: Variables 1001-1082 consist of the specific instances for group prediction pairs 2, 3, and 4 (indices _2, _3, _4), continuing the same structure

duration_group_predic: Duration in seconds spent on group prediction section

list_group: Which list group respondent was assigned to (for list experiment)

list_order_first: Order of list items shown first

max_1: Maximum value for list 1

max_2: Maximum value for list 2

list_repeat_count: Number of list experiment repetitions

list_1_answer_[1-2]: Response to list 1 in repetition [1-2]. Pattern repeats for up to 2 repetitions

list_2_answer_[1-2]: Response to list 2 in repetition [1-2]. Pattern repeats for up to 2 repetitions

duration_list: Duration in seconds spent on list experiment section

salience_1: Salience question 1 about transgender issues

salience_2: Salience question 2 about transgender issues

duration_salience: Duration in seconds spent on salience questions

m1: Follow-up question about delivery or worker preferences

m3.1: Follow-up question about delivery or worker preferences

m3.4: Follow-up question about delivery or worker preferences

m3.5: Follow-up question about delivery or worker preferences

duration_delivery_c: Duration in seconds spent on delivery follow-up section

purpose_2: Second perceived purpose question: What do you think was the main purpose of this survey?

purpose_2_other: Text entry for other perceived purposes if -97 (Other) is selected

duration_perceived_purpose_2: Duration in seconds spent on second perceived purpose question

duration_total_no_obs: Total survey duration in seconds (excluding observation sections)

hiring_obs1: Enumerator observation: Did respondent seem to understand hiring task? Options: (from questionnaire)

hiring_obs1_other: Text entry for other observations if -97 (Other) selected

hiring_obs2: Enumerator observation: How engaged was respondent in hiring task?

hiring_obs3: Enumerator observation: Did respondent discuss choices with anyone?

hiring_obs4: Enumerator observation: Did respondent seem rushed?

hiring_obs5: Enumerator observation: Any other notable behaviors during hiring task?

survey_status: Final status of survey. Options: 1=Completed, 2=Partially completed, 3=Refused

survey_status_other: Text entry for other status if -97 (Other) selected

survey_completed: Indicator of whether survey was fully completed (1=yes, 0=no)

survey_partial_reason: Reason survey was only partially completed. Options: (from questionnaire)

survey_partial_reason_other: Text entry for other partial completion reason

survey_refusal_reason: Reason respondent refused to complete survey. Options: (from questionnaire)

survey_refusal_reason_other: Text entry for other refusal reason

partial_response_sections: Which sections were completed before survey was terminated

partial_response: Indicator of whether this is a partial response (1=yes, 0=no)

video_seen: Indicator of whether respondent saw the video (1=yes, 0=no)

survey_understanding: Enumerator assessment of respondent's understanding of survey. Options: 1=Understood well, 2=Some difficulty, 3=Did not understand

survey_understanding_sections: Sections where understanding issues were noted

survey_fatigue: Enumerator assessment of respondent fatigue. Options: 1=Not fatigued, 2=Somewhat fatigued, 3=Very fatigued

survey_diff_reading: Did respondent have difficulty reading? Options: 1=Yes, 0=No

survey_genuine: Enumerator assessment of whether responses seemed genuine. Options: 1=Yes, 2=Some doubt, 3=Not genuine

survey_genuine_sections: Sections where response genuineness was questionable

survey_difficulty: Did respondent have difficulty with any part of survey? Options: 1=Yes, 0=No

survey_difficulty_type: Type of difficulty encountered. Options: (from questionnaire)

survey_difficulty_type_other: Text entry for other difficulty type

survey_difficulty_sections: Sections where difficulty was encountered

other_present: Was anyone else present during the interview? Options: 1=Yes, 0=No

other_present_who: Who else was present? Options: (from questionnaire)

other_present_who_other: Text entry for other person present

other_present_notes: Notes about others present during interview

gps-Latitude: GPS latitude coordinate of interview location

gps-Longitude: GPS longitude coordinate of interview location

gps-Altitude: GPS altitude of interview location

gps-Accuracy: Accuracy of GPS reading in meters

other_hints: Other hints or notes from enumerator

comments: General enumerator comments about the interview

duration_total_with_obs: Total survey duration in seconds (including observation sections)

instanceID: Unique identifier for this survey instance (ODK/SurveyCTO)

instanceName: Name of this survey instance

formdef_version: Version of the survey form definition used

KEY: Unique key identifier for the survey record

c13: Video recall question (nested path variable)

meta[1]/instanceName: Metadata instance name (nested path variable)

law_audio_audit: Audio file for law-related questions (nested path variable)

location: Combined location identifier

================================================================================
GRI Phase 2_WIDE.csv (Phase 2 Main Survey)
================================================================================

NOTE: Phase 2 follows the same survey structure as Phase 1, with most variables having identical definitions. The Phase 2 dataset contains 2,704 variables. Variables documented above for Phase 1 (GRI Main v2_WIDE.csv) apply to Phase 2 with the same definitions. Key differences and additional variables unique to Phase 2 are documented below.

piloting_yn: Indicator of whether this was a piloting/testing session (1=yes, 0=no)

================================================================================
NEW VARIABLES IN PHASE 2 (Not in Phase 1)
================================================================================

This section documents variables that appear in Phase 2 but not in Phase 1 of the main survey. Phase 2 introduced several experimental design enhancements: (1) an anonymous pickup method to measure revealed preferences without social desirability bias, (2) experimental variation in announcement timing (before vs. after group discussion), (3) a "choosing only" condition where participants make silent individual choices without discussion, (4) enhanced leadership scales to measure group dynamics more precisely, and (5) calculated hiring score variables to quantify discrimination patterns.

ANONYMOUS PICKUP CHOICES SECTION:
This section implements a novel method to elicit private preferences that may not be revealed in public settings. Respondents make a series of choices between delivery workers using an anonymous selection method where their choices are concealed from neighbors and observers. The section includes comprehension checks to ensure respondents understand the anonymity mechanism.

an0: Respondent selection to start anonymous pickup choices section. Options: space=Start, 0=Pass
an0_label: Label for an0 selection
an1: Anonymous pickup choice 1 (from 6 options provided)
an2: Anonymous pickup choice 2 (from 6 options provided)
an3: Anonymous pickup choice 3 (from 6 options provided)
an4: Anonymous pickup choice 4 (from 6 options provided)
an5: Anonymous pickup choice 5 (from 6 options provided)
an6: Anonymous pickup choice 6 (from 6 options provided)
an_check: Enumerator check: Has the respondent finished the anonymous pickup round? Options: 1=Yes, 0=No, -99=Refused
an_check1: Will your neighbours know which options you picked in the anonymous round? Options: 1=Yes, 0=No, -98=Don't know, -99=Refuse to respond
an_check2: Will the surveyor know which options you picked in the anonymous round? Options: 1=Yes, 0=No, -98=Don't know, -99=Refuse to respond
an_check_1: Anonymous pickup comprehension check 1
an_check_1_label: Label for anonymous pickup check 1
an_check_2: Anonymous pickup comprehension check 2 (integer)
an_check_2_label: Label for anonymous pickup check 2
an_check_3: Enumerator check for anonymous pickup completion
an_check_label: Label for anonymous pickup check
an_choice_[1-4]: Choice shown to respondent in anonymous pickup pair [1-4]. Pattern repeats for 4 pairs
an_choice_label_[1-4]: Label for anonymous pickup choice [1-4]. Pattern repeats for 4 pairs

ANONYMOUS PICKUP COMPARISON DATA:
These variables store data used to test respondents' comprehension of what other group members chose during the anonymous pickup section. This serves as a manipulation check to verify that the anonymity mechanism was effective—respondents should not be able to accurately report what others selected. Variables are indexed [1-4] for each of the 4 comparison pairs.

an_comp_choice_[1-4]: Anonymous pickup comparison choice for pair [1-4]. Pattern repeats for 4 pairs
an_comp_choice_label_[1-4]: Label for comparison choice [1-4]. Pattern repeats for 4 pairs
an_comp_correct_answer_[1-4]: Correct answer for anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_debug_[1-4]: Debug variable for anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_educ_1_[1-4]: Education level of first worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_educ_2_[1-4]: Education level of second worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_english_1_[1-4]: English proficiency of first worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_english_2_[1-4]: English proficiency of second worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_i_[1-4]: Loop index for anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_items_1_[1-4]: Items for first worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_items_2_[1-4]: Items for second worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_photo_1_[1-4]: Photo of first worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_photo_2_[1-4]: Photo of second worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_r1_choice_other1_[1-4]: Choice by first other group member in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_r1_choice_other2_[1-4]: Choice by second other group member in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_reliability_1_[1-4]: Reliability score of first worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_reliability_2_[1-4]: Reliability score of second worker in anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_rep_count: Number of repetitions in anonymous pickup comprehension check
an_comp_show_[1-4]: Display option for anonymous pickup comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_show_label_[1-4]: Label for display option in comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp_trans_1_[1-4]: Indicator if first worker in comprehension pair [1-4] is transgender. Pattern repeats for 4 pairs
an_comp_trans_2_[1-4]: Indicator if second worker in comprehension pair [1-4] is transgender. Pattern repeats for 4 pairs

SECOND SET OF ANONYMOUS PICKUP COMPREHENSION (an_comp2_*):
This is a duplicate set of the anonymous pickup comprehension check variables, following the same structure as an_comp_* above. The second round helps confirm the robustness of the anonymity manipulation by testing comprehension again with a different set of pairs.

an_comp2_choice_[1-4]: Second anonymous pickup comparison choice for pair [1-4]. Pattern repeats for 4 pairs
an_comp2_choice_label_[1-4]: Label for second comparison choice [1-4]. Pattern repeats for 4 pairs
an_comp2_correct_answer_[1-4]: Correct answer for second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_debug_[1-4]: Debug variable for second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_educ_1_[1-4]: Education level of first worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_educ_2_[1-4]: Education level of second worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_english_1_[1-4]: English proficiency of first worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_english_2_[1-4]: English proficiency of second worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_i_[1-4]: Loop index for second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_items_1_[1-4]: Items for first worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_items_2_[1-4]: Items for second worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_photo_1_[1-4]: Photo of first worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_photo_2_[1-4]: Photo of second worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_r1_choice_other1_[1-4]: Choice by first other group member in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_r1_choice_other2_[1-4]: Choice by second other group member in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_reliability_1_[1-4]: Reliability score of first worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_reliability_2_[1-4]: Reliability score of second worker in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_rep_count: Number of repetitions in second anonymous pickup comprehension check
an_comp2_show_[1-4]: Display option for second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_show_label_[1-4]: Label for display option in second comprehension pair [1-4]. Pattern repeats for 4 pairs
an_comp2_trans_1_[1-4]: Indicator if first worker in second comprehension pair [1-4] is transgender. Pattern repeats for 4 pairs
an_comp2_trans_2_[1-4]: Indicator if second worker in second comprehension pair [1-4] is transgender. Pattern repeats for 4 pairs

ANNOUNCEMENT VARIABLES:
These variables implement an experimental treatment that varies whether participants' choices are publicly announced before or after the group decision-making process. The announcement timing manipulation tests whether the salience of social accountability affects discrimination in hiring decisions. "Before" announcement makes social consequences salient before the choice; "after" announcement reduces this pressure.

announce_after: Indicator for whether choices are announced after the decision (1=yes, 0=no)
announce_before: Indicator for whether choices are announced before the decision (1=yes, 0=no)
announce_check_private: Confirmation that respondent is in a private location for announcement. Options: 1=Yes, 0=No
announce_check_private_label: Label for private location check
announce_debug_[1-4]: Debug variable for announcement pair [1-4]. Pattern repeats for 4 pairs
announce_repeat_count: Number of announcement pairs to show (4)
announce_repeat_i_[1-4]: Loop index for announcement pair [1-4]. Pattern repeats for 4 pairs

ANNOUNCEMENT REPEAT 2 VARIABLES (After group choices):
announce2_debug_[1-4]: Debug variable for second announcement round pair [1-4]. Pattern repeats for 4 pairs
announce2_repeat_count: Number of pairs in second announcement round (4)
announce2_repeat_i_[1-4]: Loop index for second announcement pair [1-4]. Pattern repeats for 4 pairs

ANONYMOUS PICKUP IMPLEMENTATION VARIABLES (anon_*):
These variables store the actual implementation details for the anonymous pickup section, including which specific delivery workers were shown in each pair and their characteristics. Unlike the comprehension check variables above, these capture the actual choices made during the anonymous selection process.

anon_i_[1-4]: Loop index for anonymous pickup pair [1-4]. Pattern repeats for 4 pairs
anon_index_1_[1-4]: Index of first worker in anonymous pickup pair [1-4]. Pattern repeats for 4 pairs
anon_index_2_[1-4]: Index of second worker in anonymous pickup pair [1-4]. Pattern repeats for 4 pairs
anon_photo_1_[1-4]: Photo of first worker in anonymous pickup pair [1-4]. Pattern repeats for 4 pairs
anon_photo_2_[1-4]: Photo of second worker in anonymous pickup pair [1-4]. Pattern repeats for 4 pairs
anon_redo: Indicator to redo anonymous pickup section (1=yes, 0=no)
anon_rep_count: Number of anonymous pickup pairs (4)
anon_trans_1_[1-4]: Indicator if first worker in anonymous pickup pair [1-4] is transgender. Pattern repeats for 4 pairs
anon_trans_2_[1-4]: Indicator if second worker in anonymous pickup pair [1-4] is transgender. Pattern repeats for 4 pairs

GROUP OBSERVATION VARIABLES - "CHOOSING ONLY" CONDITION:
These variables capture observations from the experimental "choosing only" condition, where group members make individual hiring decisions silently (without discussion) but in the presence of others. This design isolates the effect of social observability from the effect of discussion/deliberation. Observers record whether participants spoke, what they said, whether others were present, and other behaviors that might indicate social influence despite the absence of formal discussion.

choosing_only: Indicator for "choosing only" condition where participants make silent choices. Options: 1=yes, 0=no
gobs_choosing0: Observer notes: Were participants able to see each others' answers on the tablet? Options: 2=Yes for multiple pairs, 1=Yes for 1 pair, 0=No, -98=Don't know
gobs_choosing1: Observer notes: Did participants speak while choosing? Options: 1=Spoke a lot, 2=Spoke a bit, 0=Did not speak, -98=Don't know
gobs_choosing_who: Observer notes: Who spoke during choosing? Options (select_multiple): 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, -98=Don't know
gobs_choosing2: Observer notes: What did participants say? Options (select_multiple): 1=Chose A/B, 2=Asked for clarification, 3=Explained why they chose A/B, -97=Other (specify), -98=Don't know
gobs_choosing3: Observer notes: Was anyone else present during hiring choices? Options: 1=Yes, 0=No
gobs_choosing3_who: Observer notes: Who was present? Options (select_multiple): 1=Husband, 2=Wife, 3=Mother, 4=Father, 5=Sister, 6=Brother, 7=Grandparent, 8=Child, 9=Other family member, 10=Neighbour, 11=Friend, -97=Other (specify)
gobs_choosing3_what: Observer notes: What did others say? Options (select_multiple): 1=They revealed what the participant chose, 2=They said whether they agreed, 3=They gave reasons for choosing A or B, 0=Nothing, -97=Other
gobs_choosing5: Observer observation for choosing round 5
gobs_choosing6: Observer observation for choosing round 6
gobs_choosing7: Observer observation for choosing round 7
gobs_choosing7_other: Text entry for other observations in choosing round 7

GROUP DISCUSSION VARIABLES - PAIRWISE OBSERVATIONS:
These variables capture detailed observations during pairwise group discussions, where two participants discuss their choices while a third acts as a "listener." This design helps identify who influences whom in the group decision-making process by tracking the listener's engagement and participation.

gobs_pair3: Observer notes: Did the listener speak while others were discussing? Options: 1=Spoke a lot, 2=Spoke a bit, 0=Did not speak, -98=Don't know
gobs_pair4: Observer notes: What did the listener say? Options (select_multiple): 1=Chose A/B, 2=Asked for clarification, 3=Explained why they chose A/B, -97=Other (specify), -98=Don't know
gobs_pair4_n: Observer notes: For how many pairs (out of 4) did the listener choose A/B or explain? (integer)
gobs_pair5: Observer notes: Did the listener listen to the discussion? Options: 1=Listened closely, 2=Listened somewhat, 0=Did not listen at all, -98=Don't know
gobs7: In general, who was most dominant in the conversation? Options: 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}, -98=Don't know

ENHANCED LEADERSHIP SCALES:
Phase 2 expanded the measurement of leadership and social influence within groups. These scales assess who holds leadership positions in the group, how confident/assertive each member is, and ratings of influence across multiple dimensions. The "_a" and "_b" suffixes typically refer to ratings of two different group members. These enhanced measures allow researchers to more precisely identify the mechanisms through which group discussions affect discrimination.

gc8: In general, do you think that [group member] is similar to you? Options: (select_one similar_to_me from questionnaire)
ls1: Out of 10, how confident is [group member]? (integer 0-10)
ls2: Out of 10, how quiet is [group member]? (integer 0-10)
ls3: Out of 10, how like a leader is [group member]? (integer 0-10)
ls6: Out of 10, how shy is [group member]? (integer 0-10)
ls7: Out of 10, how talkative is [group member]? (integer 0-10)
ls8: Out of 10, how admirable is [group member]? (integer 0-10)
ls9: Out of 10, how inspiring is [group member]? (integer 0-10)

LEADERSHIP SCALE (SELF-REPORT):
ls_self1: Self-reported leadership scale item 1 (integer 0-10)
duration_leadership_scale_self: Duration in seconds spent on self-report leadership scale

DETAILED PAIR OBSERVATIONS (gobsv2_*):
These variables provide more granular observation data for each hiring pair discussion, going beyond the basic observations collected in Phase 1. They capture who initiated conversations, who advocated for or against transgender workers, the specific reasons articulated, and other nuanced aspects of the group deliberation process. This enhanced data collection enables fine-grained analysis of how social influence operates in groups.

gobsv2_1: Who spoke first in this pair's discussion? Options vary by discussion type
gobsv2_3: Who spoke in favour of transgenders? Options (select_multiple) vary by discussion type
gobsv2_6: What were the main reasons for the choice? Options (select_multiple) vary by discussion type
gobsv2_7: Additional observation variable 7 for group pair
gobsv2_8: Additional observation variable 8 for group pair
gobsv2_9: Additional observation variable 9 for group pair
gobsv2_10: Additional observation variable 10 for group pair
gobsv2_as_1: Group choice observation: specific attribute 1
gobsv2_as_2: Group choice observation: specific attribute 2
gobsv2_as_3: Group choice observation: specific attribute 3
gobsv2_as_4: Group choice observation: specific attribute 4
gobsv2_as_4_pair: Group choice observation: pair-specific attribute 4
gobsv2_as_5: Group choice observation: specific attribute 5
gobsv2_as_5_other: Text entry for "other" in group choice observation attribute 5
gobsv2_as_9: Group choice observation: specific attribute 9
gobsv2_as_10: Group choice observation: specific attribute 10

HIRING VARIABLES - EXPANDED:
check_group_pair: Enumerator check: Which participants are in the discussion? Options (select_multiple): 1=${group_person_1}, 2=${group_person_2}, 3=${group_person_3}
check_group_pair_label: Label for group pair check
check_private_ind: Confirmation that individual is in private area. Options: 1=Yes, 0=No
hiring_selected_group_choice_speaker_1: The first speaker selected by the group for hiring
hiring_selected_group_choice_speaker_2: The second speaker selected by the group for hiring
hiring_selected_group_choice_speaker_backup: Backup speaker selected by the group for hiring

CALCULATED HIRING SCORE VARIABLES:
Phase 2 introduced systematic calculation of discrimination scores based on hiring choices. These calculated variables quantify the degree of discrimination by modeling hiring decisions as a function of worker characteristics (education, reliability, English proficiency) and transgender status. The coefficients, predictions, and scores enable researchers to measure discrimination on a continuous scale rather than relying solely on binary choice comparisons.

check_private: Confirmation of private location for hiring round. Options: 1=Yes, 0=No
coeff_item_diff: Coefficient for item difference (calculated variable)
coeff_r2_reliability_diff: Coefficient for reliability difference (calculated variable)
coeff_r2_reliability_shown: Coefficient for reliability shown (calculated variable)
hiring_choice_reason: Reason for hiring choice (from questionnaire options)
intercept: Intercept value for hiring model (calculated variable)
predic_log_odds: Predicted log odds for hiring choice (calculated variable)
predic_p: Predicted probability for hiring choice (calculated variable)
score_pair_bonus: Bonus score for the pair (calculated variable)
score_pair_bonus_minus: Negative bonus score adjustment (calculated variable)
score_pair_bonus_plus: Positive bonus score adjustment (calculated variable)
score_pair_only_trans: Score when only transgender option in pair (calculated variable)
score_pair_raw: Raw score for the pair (calculated variable)
total_score: Total score across all hiring decisions (calculated variable)
total_score_100: Total score scaled to 100 (calculated variable)

GROUP DISCUSSION IMPLEMENTATION VARIABLES:
These variables manage the implementation of the group discussion condition, including tracking which participants served as "speakers" (those whose choices were discussed) versus "listeners." The speaker/listener design allows identification of peer influence effects by comparing how choices differ when made privately versus when made after hearing others' reasoning.

group_discussion_yn: Indicator for whether there was a group discussion. Options: 1=Yes, 0=No
group_round_piloting_choices: Piloting variable for group round choices
group_take_from_group_choice_speaker_1: First speaker choice taken from group discussion
group_take_from_group_choice_speaker_2: Second speaker choice taken from group discussion
group_take_from_group_choice_speaker_backup: Backup speaker choice taken from group discussion
group_choice_speaker_1: First speaker chosen by the group
group_choice_speaker_2: Second speaker chosen by the group
group_choice_speaker_backup: Backup speaker for the group
speaker_1_choice_[1-4]: Choice made by speaker 1 for pair [1-4]. Pattern repeats for 4 pairs
speaker_2_choice_[1-4]: Choice made by speaker 2 for pair [1-4]. Pattern repeats for 4 pairs
r1_choice_speaker_1: Round 1 choice by speaker 1
r1_choice_speaker_1_num: Numeric code for round 1 choice by speaker 1
r1_choice_speaker_2: Round 1 choice by speaker 2
r1_choice_speaker_2_num: Numeric code for round 1 choice by speaker 2
group_repeat2_count: Number of second-level group repetitions
group_repeat2_i: Loop index for second-level group repetition

ADDITIONAL DELIVERY VARIABLES:
delivery_incentive_yn: Yes/no indicator for delivery incentive. Options: 1=Yes, 0=No
demog: Demographics indicator variable

MEMORY/RECALL MANIPULATION CHECKS:
These variables test whether participants can accurately recall what others chose or what they themselves chose in various experimental conditions. Memory checks serve as manipulation checks: if the design successfully creates private vs. public conditions, participants should have better recall in public (discussion) conditions than in private (choosing only or anonymous) conditions.

memcheck_choice_choosing_only_show: Display of options for memory check in choosing only condition
memcheck_choice_choosing_only: Memory check: Which did the other participants select? Options: 1=Both picked A, 2=One picked A one picked B, 3=Both picked B, -98=Don't know, -99=Refused
memcheck_choice_d_show: Display of options for memory check (discussion)
memcheck_choice_d: Memory check for discussion: Which did others pick? Options: 1=Others picked A, 2=Others picked B, -98=Don't know, -99=Refused
memcheck_choice_d_speak_show: Display of options for speaker memory check
memcheck_choice_d_speak: Memory check: Which did you pick in group discussion? Options: 1=We picked A, 2=We picked B, -98=Don't know, -99=Refused
memcheck_reason: Memory check: Why did other participants say they selected this option? Options (select_multiple) same as hiring reasons

REASONS FOR CHOICES (PAIR-LEVEL):
pair_reason_count: Number of pairs for which reasons are collected
f2_show: Display of the choice made for reason follow-up. Options: 1=Person A, 2=Person B
f2: Why did you choose this option? Options (select_multiple) same as hiring reasons (see hiring reason variables above)

DURATION VARIABLES (NEW):
duration_scratchcard_choice: Duration in seconds for scratchcard selection

DEBUG VARIABLES:
debug: General debug variable
debug_sc: Debug variable for scratchcard selection

NOTE: All other Phase 2 variables follow the same structure and definitions as Phase 1. The larger variable count (2,704 vs 1,181) is primarily due to: (1) the addition of the anonymous pickup section with comprehension checks, (2) expanded announcement timing variables, (3) enhanced leadership measurement scales, (4) the "choosing only" experimental condition with detailed observation variables, (5) calculated hiring score variables for quantifying discrimination, and (6) memory/recall manipulation checks for validating experimental conditions.

================================================================================
GRI Follow Up Delivery v3_WIDE.csv (Follow-up Survey)
================================================================================

This is the follow-up survey administered after participants received their actual delivery from the worker they selected in the main survey. The follow-up captures: (1) verification that the delivery occurred and matches records, (2) participants' experience with and reactions to the delivery worker, (3) post-intervention attitudes and hiring preferences to measure attitude change, and (4) additional hiring decisions with new worker pairs to assess behavioral persistence. The follow-up contains 368 variables, with many variables reused from Phase 1/2 (demographics, hiring decisions, social norms, knowledge questions) having identical definitions.

KEY FOLLOW-UP SPECIFIC VARIABLES:

SURVEY VERIFICATION AND TRACKING:
These variables confirm the respondent's identity and participation in the original survey, ensuring data can be properly linked across survey waves.

respondent_availability: Was respondent available for follow-up? Options: 1=Yes, 0=No

respondent_unavailable_reason: Reason respondent was unavailable. Options: (from questionnaire)

respondent_unavailable_reason_other: Text entry for other unavailability reason

name_confirm: Confirmation that respondent name matches records. Options: 1=Yes, 0=No

name_redo_text: Corrected name if original was incorrect

name_redo_why: Reason for name correction. Options: (from questionnaire)

name_redo_why_other: Text entry for other name correction reason

first_survey_date: Date when the first (main) survey was conducted

confirm_first_survey: Confirmation that respondent participated in first survey. Options: 1=Yes, 0=No

confirm_first_survey_fail_reason: Reason for failure to confirm first survey. Options: (from questionnaire)

confirm_first_survey_fail_reason_other: Text entry for other confirmation failure reason

first_survey_date_correction: Corrected date if original was incorrect

confirm_first_survey_in_hh: Confirmation that first survey was in this household. Options: 1=Yes, 0=No

confirm_first_survey_comment: Comments about first survey confirmation

PRE-DELIVERY QUESTIONS:
These questions are asked before discussing the delivery experience, capturing baseline expectations and any pre-intervention attitudes to distinguish from post-delivery effects.

p1: Pre-delivery question 1

p2: Pre-delivery question 2

p3: Pre-delivery question 3

p3_other: Text entry for other response to p3

p4: Pre-delivery question 4

duration_pre_qs: Duration in seconds spent on pre-delivery questions

DELIVERY VERIFICATION AND CONSENT:
These variables confirm that the delivery occurred as planned and obtain consent to discuss the delivery experience. Non-delivery or refusal to discuss the delivery is also documented.

confirm_deliv: Confirmation of delivery details. Options: 1=Yes, 0=No

consent_pre_deliv: Consent for pre-delivery questions. Options: 1=Yes, 0=No

deliv_consent_refused: Reason for refusing delivery consent. Options: (from questionnaire)

deliv_consent_refused_other: Text entry for other refusal reason

duration_ask_deliv: Duration in seconds spent asking about delivery

DELIVERY EXPERIENCE QUESTIONS:
These questions capture the respondent's experience with the actual delivery, including satisfaction, interactions with the worker, and any notable events or reactions. These are the key outcome measures for assessing whether contact with transgender workers changes attitudes.

d1: Delivery-related question 1

d2: Delivery-related question 2

d2_other: Text entry for other response to d2

d3: Delivery-related question 3

duration_delivery_qs: Duration in seconds spent on delivery questions

FOLLOW-UP HIRING DECISIONS:
After discussing the delivery experience, respondents make a new set of hiring decisions with different worker pairs. This measures whether the delivery experience changed hiring behavior. Comparing these choices to the original survey choices identifies treatment effects of contact with transgender workers. The structure follows the same pattern as the main survey hiring variables, with pairs indexed [1-6].

follow_up_i_[1-6]: Loop index for follow-up hiring decision pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_index_1_[1-6]: Index of first delivery worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_index_2_[1-6]: Index of second delivery worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_photo_1_[1-6]: Photo filename for first worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_reliability_1_[1-6]: Reliability score of first worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_items_1_[1-6]: Items to be delivered by first worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_trans_1_[1-6]: Indicator if first worker in follow-up pair [1-6] is transgender (1=transgender, 0=not). Pattern repeats for up to 6 pairs

follow_up_photo_2_[1-6]: Photo filename for second worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_reliability_2_[1-6]: Reliability score of second worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_items_2_[1-6]: Items to be delivered by second worker in follow-up pair [1-6]. Pattern repeats for up to 6 pairs

follow_up_trans_2_[1-6]: Indicator if second worker in follow-up pair [1-6] is transgender (1=transgender, 0=not). Pattern repeats for up to 6 pairs

FOLLOW-UP GROUP PREDICTIONS:
For respondents who participated in groups, the follow-up includes predictions about what other group members would choose. This measures whether the delivery experience changed perceptions of peers' attitudes, which could indicate shifts in perceived social norms. Variables follow the same structure as the main survey prediction variables, with pairs indexed [1-4].

group_predic_fu_i_[1-4]: Loop index for group prediction pair [1-4] in follow-up. Pattern repeats for up to 4 pairs

group_predic_fu_index_1_[1-4]: Index of first delivery worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_index_2_[1-4]: Index of second delivery worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_photo_1_[1-4]: Photo filename for first worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_age_1_[1-4]: Age of first worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_educ_1_[1-4]: Education level of first worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_reliability_1_[1-4]: Reliability score of first worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_english_1_[1-4]: English proficiency of first worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_items_1_[1-4]: Items to be delivered by first worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_trans_1_[1-4]: Indicator if first worker in follow-up prediction pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

group_predic_fu_photo_2_[1-4]: Photo filename for second worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_age_2_[1-4]: Age of second worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_educ_2_[1-4]: Education level of second worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_reliability_2_[1-4]: Reliability score of second worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_english_2_[1-4]: English proficiency of second worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_items_2_[1-4]: Items to be delivered by second worker in follow-up prediction pair [1-4]. Pattern repeats for up to 4 pairs

group_predic_fu_trans_2_[1-4]: Indicator if second worker in follow-up prediction pair [1-4] is transgender (1=transgender, 0=not). Pattern repeats for up to 4 pairs

DISCUSSION AND DEBRIEF QUESTIONS:
These open-ended and structured questions elicit detailed accounts of the delivery experience, reasons for any attitude changes, and reflections on the intervention. They provide qualitative context to interpret quantitative changes in hiring behavior.

discuss1: Discussion question 1 about delivery experience

discuss3: Discussion question 3 about delivery experience

discuss3.1: Discussion sub-question 3.1

discuss3.1_other: Text entry for other response to discuss3.1

discuss3.2: Discussion sub-question 3.2

discuss3.2_other: Text entry for other response to discuss3.2

discuss3.3: Discussion sub-question 3.3

discuss3.4: Discussion sub-question 3.4

discuss3.5: Discussion sub-question 3.5

discuss3.6: Discussion sub-question 3.6

duration_field_discussion: Duration in seconds spent on field discussion questions

duration_multi_delivery_organise: Duration in seconds spent organizing multiple deliveries

GPS AND METADATA:
Standard survey metadata variables for tracking interview location and data quality in the follow-up survey.

gps_follow_up-Latitude: GPS latitude of follow-up interview location

gps_follow_up-Longitude: GPS longitude of follow-up interview location

gps_follow_up-Altitude: GPS altitude of follow-up interview location

gps_follow_up-Accuracy: Accuracy of follow-up GPS reading in meters

NOTE: The follow-up survey also includes many variables from the main survey that are repeated to measure change: social norms questions (n1-n3), knowledge/awareness questions (k1-k11), beliefs about transgender rights, salience measures, and survey quality observations. These reused variables follow the same definitions as documented in the main survey sections above.

================================================================================
GRI Follow Up Delivery - Phase 2_WIDE.csv (Follow-up Survey Phase 2)
================================================================================

This is the Phase 2 follow-up survey administered after the delivery intervention. It contains 269 variables and follows the same structure as the Phase 1 follow-up survey (GRI Follow Up Delivery v3_WIDE.csv), with all variables having identical definitions. The Phase 2 follow-up has fewer variables than Phase 1 follow-up due to streamlining of some sections based on Phase 1 experience.

ADDITIONAL PHASE 2 FOLLOW-UP VARIABLES:

back_check_yn: Indicator of whether this was a back-check/audit interview (1=yes, 0=no)
back_check_rand: Random assignment for back-check selection

NOTE: All other variables follow the definitions in the Phase 1 follow-up documentation above.

================================================================================
END OF CODEBOOK
================================================================================
